### Current state
We have FilterChooser class, which can be thought of as a **tree of
encodings**. Tree nodes are instances of FilterChooser itself, and come
in two types:
* A node containing single encoding that has *constant* bits in the
specified bit range, a.k.a. singleton node.
* A node containing only child nodes, where each child represents a set
of encodings that have the same *constant* bits in the specified bit
range.
Either of these nodes can have an additional child, which represents a
set of encodings that have some *unknown* bits in the same bit range.
As can be seen, the **data structure is very high level**.
The encoding tree represented by FilterChooser is then converted into a
finite-state machine (FSM), represented as **byte array**. The
translation is straightforward: for each node of the tree we emit a
sequence of opcodes that check encoding bits and predicates for each
encoding. For a singleton node we also emit a terminal "decode" opcode.
The translation is done in one go, and this has negative consequences:
* We miss optimization opportunities.
* We have to use "fixups" when encoding transitions in the FSM since we
don't know the size of the data we want to jump over in advance. We have
to emit the data first and then fix up the location of the jump. This
means the fixup size has to be large enough to encode the longest jump,
so **most of the transitions are encoded inefficiently**.
* Finally, when converting the FSM into human readable form, we have to
**decode the byte array we've just emitted**. This is also done in one
go, so we **can't do any pretty printing**.
### This PR
We introduce an intermediary data structure, decoder tree, that can be
thought as **AST of the decoder program**.
This data structure is **low level** and as such allows for optimization
and analysis.
It resolves all the issues listed above. We now can:
* Emit more optimal opcode sequences.
* Compute the size of the data to be emitted in advance, avoiding
fixups.
* Do pretty printing.
Serialization is done by a new class, DecoderTableEmitter, which
converts the AST into a FSM in **textual form**, streamed right into the
output file.
### Results
* The new approach immediately resulted in 12% total table size savings
across all in-tree targets, without implementing any optimizations on
the AST. Many tables observe ~20% size reduction.
* The generated file is much more readable.
* The implementation is arguably simpler and more straightforward (the
diff is only +150~200 lines, which feels rather small for the benefits
the change gives).
`Predicates` and `Features` fields serve the same purpose. They should
be kept in sync, but not all predicates are based on features. This
resulted in introducing dummy features for that only reason.
This patch removes `Features` field and changes TableGen emitters to use
`Predicates` instead.
Historically, predicates were written with the assumption that the
checking code will be used in `SelectionDAGISel` subclasses, meaning
they will have access to the subclass variables, such as `Subtarget`.
There are no such variables in the generated
`GenSubtargetInfo::getHwModeSet()`, so we need to provide them. This can
be achieved by subclassing `HwModePredicateProlog`, see an example in
`Hexagon.td`.
It can never be reached. It could be reached if we emitted an opcode
that could fall outside the outermost scope, but emission of all such
opcodes is guarded by `!isOutermostScope()`.
That also means we never add fixups to the outermost scope, so avoid
pushing an entry for it onto the stack.
So we can see the changes in table sizes after making changes to
DecoderEmitter by simply running `grep DecoderTable`.
Also, remove an unnecessary terminating 0 from the end of the tables.
Previously, HW mode name was appended to decoder namespace name when
enumerating encodings, and then emitTable appended the bit width to it
to form the final table name. Let's do this all in one place.
A nice side effect is that this allows us to avoid having to deal with
std::string.
The changes in the tests are caused by the different order of tables.
1. Remove 'AllModes' and 'DefaultMode' suffixes for DecoderTables under
default HwMode.
2. Introduce a less aggressive suppression for HwMode DecoderTable, only
reduce necessary tables duplications. This allows encodings under
different HwModes to retain the original DecoderNamespace.
3. Change 'suppress-per-hwmode-duplicates' command option from bool type
to enum type, allowing users to choose what level of suppression to use.
Currently, for per-HwMode encoding/decoding, those instructions that do
not have a HwMode override are duplicated into the decoder tables for
all HwModes. This includes inducing multiple tables for instructions
that are otherwise unrelated (e.g., different namespace with no
overrides at all).
This patch adds support to suppress instruction and table duplicates.
TableGen option "-gen-disassembler --suppress-per-hwmode-duplicates"
enables the suppression (off by default).
For one downstream backend with a complicated ISA and major
cross-generation encoding differences, this eliminates ~32000 duplicate
table entries at the time of this patch.
There are legitimate reasons to suppress or not suppress duplicates. If
there are relatively few non-overridden related instructions, it can be
convenient to pull them into the per-mode tables (only need to decode
the per-mode tables, slightly simpler decode function in disassembler).
On the other hand, in some backends, the opposite is true or the size is
too large to tolerate any duplication in the first place. We let the
user decide which makes sense.
This is currently off by default, though there is no reason it couldn't
be enabled by default. Any existing backends downstream using the
per-HwMode feature will function as before. Turning on the feature
requires minor modifications to their disassembler due to more/less
tables and naming.
